Thư viện prompt AI OpenZhenZhen · MiniMax H3 · video

Chiến dịch truy đuổi tại quảng trường

Plaza Pursuit Operation

Chiến dịch truy đuổi tại quảng trường — khám phá tác phẩm được tạo bằng MiniMax H3 trên OpenZhenZhen, kèm prompt gốc và ví dụ tạo nội dung để tham khảo.

Chiến dịch truy đuổi tại quảng trường — AI video creation
Chiến dịch truy đuổi tại quảng trường

Prompt

Prompt tiếng Trung

MiniMax H3 视频生成规格。时长:严格 20 秒。画幅:16:9。制作一段现代城市办公楼外广场的写实电影感日间动作场景。使用所提供的身份参考图确定女特工韩梅的面部外观;她始终穿全黑服装并背肩包。镜头1(0–5秒):中景缓慢向右摇摄。戴墨镜、穿西装的特工李雷蹲在路边掩护。韩梅触碰右耳通讯器,目光锐利,用低沉克制的声音说:“我现在盯着他们。”身旁穿黑色T恤的张华神情严肃、保持警戒。用浅景深突出人物表情。镜头2(5–9秒):局势突然变化。韩梅向前冲刺。使用甩镜、快速移焦、拉远至广角,并加入克制的手持晃动。远处背景中,穿反光背心的嫌疑人王强正与穿灰衣的赵伟激烈扭打。镜头3(9–15秒):在广场绿化带旁,韩梅借助冲势跃起,将逃跑的嫌疑人扑倒在地。张华紧跟冲入画面,制服另一名目标。采用低机位和动态斯坦尼康跟拍,动作清晰有力且符合物理逻辑。镜头4(15–20秒):张华单膝跪地牢牢压住目标,严厉说道:“不许动!”韩梅迅速蹲到旁边协助控制。先以面部特写开始,随制服动作向下倾斜,最后稳定为平视固定镜头,采用三分法构图。保持人物身份一致、自然日光、写实编排、清晰空间连续性和电影化声音设计。不得使用任何源视频或源音频作为输入。

Prompt tiếng Anh

MiniMax H3 video generation specification. Duration: exactly 20 seconds. Aspect ratio: 16:9. Create a realistic cinematic daytime action scene outside a modern city office building. Use the supplied identity reference image for Agent Han Mei’s facial appearance; keep her dressed entirely in black with a shoulder bag. Shot 1 (0–5s): Medium shot with a slow pan right. Suit-and-sunglasses agent Li Lei crouches roadside as cover. Han Mei touches the communicator in her right ear, watches sharply, and says in a low, controlled voice, “I’m watching them now.” Zhang Hua, wearing a black T-shirt, stays solemn and alert beside her. Use shallow depth of field to emphasize expressions. Shot 2 (5–9s): The situation changes abruptly. Han Mei sprints forward. Use a whip pan, rapid focus shift, pull back to a wide shot, and restrained handheld shake. In the distant background, suspect Wang Qiang in a reflective vest struggles with Zhao Wei, a man in gray. Shot 3 (9–15s): Along the landscaped plaza strip, Han Mei leaps and tackles the fleeing suspect to the ground. Zhang Hua rushes into frame and subdues another target. Use a low angle and dynamic Steadicam tracking, with clear, impactful, physically coherent action. Shot 4 (15–20s): Zhang Hua drops to one knee and pins the target, firmly saying, “Don’t move!” Han Mei crouches beside him to assist. Begin with a close-up, tilt down with the restraint, then settle into an eye-level locked-off shot using rule-of-thirds composition. Maintain consistent character identities, natural daylight, realistic choreography, crisp spatial continuity, and cinematic sound design. Do not use any source video or source audio as input.

Chủ đề và phong cách

Xem mô hình API

Xem thêm prompt hình ảnh và video · Tạo hình ảnh · Tạo video · Tạo trên canvas AI